Opti-U

A performance digital twin for every player.

Opti-U turns each player's telemetry into a model of how that player trains, recovers and performs. Machine learning finds the patterns, forecasts the days ahead and models injury risk and burnout, so staff can act early.

Talk to us

Product

Inside the dashboard.

Each discipline gets its own portal, opening on availability, forecasts and the players who changed since yesterday. These screens come from our design prototype: the club, the players and every reading are fictional.

Squad availability for a fictional club: 1,790 availability days, 93.1% of roster days, beside a projection of available players over the next 14 days with its range
Squad availability
Matchday availability two days before an away league fixture: 17 athletes available, 2 limited and 1 unavailable, each with the date the medical team expects them back
Matchday
Physical performance against each athlete's own baseline, with the latest total distance, the baseline, a z-score, the smallest detectable change and whether the change is beyond noise
Own baselines
A logged restriction decision listing the options considered, the option selected, the snapshots it rests on and the physiotherapist's sign-off
Decision log

About

Every player gets a digital twin.

Each twin is a model of one player, updated with every session, night's sleep and check-in. It sits above the GPS and athlete management systems a club already runs, and every value carries its confidence, data sufficiency and source.

The twin

From raw telemetry to the next call.

  1. <01>

    Data telemetry

    GPS, training load, wearables, match data and daily check-ins flow into one record for each player, with the source of every reading kept.

  2. <02>

    Machine learning analysis

    Models learn each player's own baselines and update as new data arrives, so every reading is judged against that player's normal.

  3. <03>

    Pattern recognition

    The twin compares today with the player's whole history, finds the periods that looked alike and shows what followed them.

  4. <04>

    Forecasting

    Sleep, HRV and resting heart rate are forecast 14 days ahead and recovery 7 days ahead, each with its range.

  5. <05>

    Injury risk and burnout modelling

    Injury, illness, physical burnout and performance dip risk are modelled for each player over the next 14 to 21 days, against the squad's usual rate.

Forecasting

See where each player is heading.

The twin projects each player's sleep, HRV and resting heart rate 14 days ahead, built on their last 28 days and set against their own baseline. Every next-day forecast is checked against the reading that follows it.

A player's twin forecasting sleep duration and HRV 14 days ahead, each with an 80% range and their own baseline

Injury risk

Injury risk for every player, every day.

The twin produces a fresh injury and illness risk forecast for each player each day. Medical staff can watch it build, compare it with the squad's usual rate and see the signals behind it.

A fictional player's injury or illness risk as forecast on each of the last 28 days, rising from the squad's usual rate to 18%, with its range

Burnout

Burnout and performance dips, modelled early.

Physical burnout and performance dip risk sit beside availability risk for every player, each with its range. Staff can see which player has moved away from the squad's usual rate and adjust the week.

Forecasts for a fictional player: availability risk 18% and physical burnout risk 2%, with performance dip risk 15%, each against the squad's usual rate

Pattern recognition

The twin remembers what came next.

When a player starts to look like an earlier period, the twin shows that period, how similar it is and what followed. Staff see the familiar run of signals while there is still time to change the plan.

Pattern memory for a fictional player: an earlier period 81% similar to today was followed by the physical burnout pattern five days later, and a period 63% similar was followed by nothing

The attention flag

An early flag on any player.

The flag rises when two or more of a player's own signals move beyond their usual range, across at least two layers such as fatigue and soreness. Staff see the signals that raised it and what to check first. On a normal day it stays down.

An attention flag raised for a fictional player as self-reported fatigue and soreness move past their own range, with seven signals charted against that range

GPS vendors record each session.

The twin reads it against that player's own normal.

Staff spot trouble after it arrives.

The twin forecasts it each morning.

Decisions stay in people's heads.

Every call is logged with what was weighed.

The Shift

The system

Athlete twin

A performance digital twin of each player that lasts across seasons, built from telemetry, availability, check-ins and practitioner decisions.

Club dashboard

A separate portal for medical, performance, sports science, coaching, analysis, nutrition, psychology and leadership. Each sees only its own access tier.

Athlete app

A daily check-in in under a minute, the player's own data, their consents and a log of who viewed their record.

Trust

Staff stay in charge of every decision.

The twin gives staff the evidence. Qualified practitioners make each significant decision and the system records it. Every read of a player's record goes into a tamper-evident log, medical and psychology records sit behind separate firewalls, and player data is handled as special category data under UK and EU GDPR.

The access log: each read is listed with its time, the role that made it, the action, the subject, the workspace and a hash in a tamper-evident chain

For players

Players see what is asked of them.

Requests from staff arrive in one place, and each says which team receives the answer. Detailed physiology from a player's own wearable stays off until they switch it on, and they can switch it off again.

The athlete's requests page with five requests due from staff, each naming who set it and which team receives the answer

Later

With the player's authorisation, their twin will be able to move with them on a transfer, so staff at the new club start from an established normal.

Contact

Opti-U is looking for design partner clubs.

Opti-U's statistical engine is built and tested on real human data. The sports layer is a working design prototype on simulated data. Design work with partner clubs starts in the 2027/28 season, and independent researchers will evaluate the first live season.

info@opti-u.com